Ordinary computers use components called transistors which use two states, on and off (bits), to represent value and complete calculations. In an emerging field the components that do that job are much smaller and offer a third state. These components are fundamental particles which have a potential for superposition and can have a spin of 1, 0 or 1 and 0 (qubits). This allows for much more complex work to be done with much fewer qubits than bits. Earlier in 2020, Google achieved quantum supremacy by employing a quantum computer to do work that no supercomputer could do. The field of quantum computing is new, there’s still lots of potential. One goal more immediate than quantum supremacy is to find a practical solution to creating a good environment for quantum computing. The particles are easily influenced by their surroundings and any small error in the environment they sit in could influence the results of whatever computation is being done.
This article is about optimizing the way robots take environmental input. When humans look at a landscape we process it accurately under far more diverse conditions than a robot can with the previous technology. The problem is that a camera compresses images using a process called tone mapping. This process usually narrows an image to a smaller range of possible pixel brightness. Researchers in the Army hope to improve the range so that more consistent images can be processed. In widening that range research is also being done in how the human brain processes images since the brain also uses a very wide range in terms of the brightness of “pixels”. This research in turn could help robots with their processing.